Sequential sparse representation for mitotic event recognition. Issue 14 (1st July 2013)

• Record Type: Journal Article
• Authors: Liu, A.A.; Hao, T.; Gao, Z.; Su, Y.T.; Yang, Z.X.
• Is Part Of: Electronics letters. Volume 49:Issue 14(2013)
• Page Start: 869
• Page End: 871
• Publication Date: 2013-07-01
• Subjects: cell motility -- image segmentation -- optical microscopy -- spatiotemporal phenomena; mitotic event recognition -- sequential sparsity representation method -- microscopy image segmentation method -- mitotic candidate extraction -- dictionary learning -- sparsity decomposition -- convex objective function -- spatiotemporal event modelling -- temporal inference model -- F1 score; cell motility -- image segmentation -- optical microscopy -- spatiotemporal phenomena; mitotic event recognition -- sequential sparsity representation method -- microscopy image segmentation method -- mitotic candidate extraction -- dictionary learning -- sparsity decomposition -- convex objective function -- spatiotemporal event modelling -- temporal inference model -- F1 score; Electronics -- Periodicals; 621.381
• DOI: 10.1049/el.2013.0197
• Languages: English
• ISSNs: 0013-5194

Abstract

Abstract : Proposed is a sequential sparsity representation method for mitotic event recognition. First, an imaging model‐based microscopy image segmentation method is implemented for mitotic candidate extraction. Then, a sequential sparsity representation scheme is proposed for dictionary learning and sparsity decomposition for sequential events. Specifically, a convex objective function jointly regularised by sparsity, consistent and smooth terms is formulated to compute the reconstructed residual, which is finally utilised for classification. This method can take advantage of temporal context for spatio‐temporal event modelling. Moverover, it can overcome the shortage of temporal inference models which highly depends on a large amount of training data and long‐range temporal context. The comparison shows that this method can outperform competing methods in terms of precision, recall and F1 score.